Wall Street is undergoing a fundamental transformation as major financial institutions rapidly deploy AI agents to work alongside human employees. Danny Goldman, cofounder of Mako AI—a Khosla Ventures-backed startup offering generative AI assistants for junior finance professionals—predicts that within one to two years, every junior banker on Wall Street will have their own AI direct report. His guidance to users is simple yet profound: “Talk to this like a teammate and treat it like a teammate.”
JPMorgan Chase is leading the charge in AI adoption at the executive level. CEO Jamie Dimon is described as a “tremendous user” of the bank’s generative AI assistant suite, according to Teresa Heitsenrether, JPMorgan’s chief data and analytics officer. Speaking at the Evident AI Symposium last week, Heitsenrether revealed that JPMorgan is developing personalized AI assistants tailored to individual employees and their specific job functions, with hopes of having a clearer implementation picture by next year.
The evolution from simple task automation to sophisticated multi-agent systems represents the next frontier. Currently, most AI agents handle discrete tasks like querying internal databases, creating PowerPoint presentations, and drafting emails. However, technologists and investors are now pushing toward multi-agent systems that coordinate multiple AI agents to autonomously complete complex, multi-step tasks. Some tech executives at the Evident AI Symposium suggested the number of AI agents could exceed the human population by 2025.
Trust and verification remain critical challenges to widespread adoption. Sumitra Ganesh from JPMorgan’s AI research team acknowledged that “we don’t have a lot of trust right now in these systems,” necessitating a careful “slow-walk” approach where experts verify AI outputs before actions are taken. She compared the current state to “babysitting these agents” with “training wheels” that will eventually come off as confidence builds.
Performance benchmarking is already underway. A recent University of Cambridge study comparing AI agents to humans in business management found that AI outperformed humans on metrics including profitability, product design, and inventory management, though humans still excelled at making real-time decisions. Gabriel Stengel, CEO of Rogo—which is building a generative AI equivalent of a junior banker—noted that the industry is “still figuring out the tasks they’re actually good at” and which tools they can effectively use.
Key Quotes
Talk to this like a teammate and treat it like a teammate.
Danny Goldman, cofounder of Mako AI, provides this guidance to private-equity customers using his startup’s generative AI assistant. This philosophy reflects the fundamental shift in how AI is being positioned—not as software tools, but as collaborative team members that require human-like interaction patterns.
We don’t have a lot of trust right now in these systems. We have to slow-walk it to release it to people who are experts who can verify the output and go, ‘Okay, that looks fine, you can take that action.’
Sumitra Ganesh, a member of JPMorgan’s AI research team, articulated the primary barrier to fully autonomous AI agents at the Evident AI Symposium. Her comments highlight the cautious approach major financial institutions are taking despite their enthusiasm for AI adoption.
What’s really exciting about agents is that we are still figuring out the tasks they’re actually good at, the tools they know how to use, the tools we have to teach them how to use.
Gabriel Stengel, CEO of Rogo, emphasized that the AI agent revolution is still in its experimental phase. His observation underscores that even as deployment accelerates, the industry is learning in real-time about optimal use cases and limitations.
But that’s kind of babysitting these agents at this point. But hopefully, it’s like training wheels — at some point, we will be confident enough to let them go.
Sumitra Ganesh from JPMorgan used this metaphor to describe the current state of AI agent supervision, suggesting that today’s intensive oversight is temporary as organizations build confidence in AI reliability and develop better guardrails for autonomous operation.
Our Take
The financial services industry’s embrace of AI agents represents more than technological adoption—it’s a fundamental reimagining of organizational structure and human capital. The most striking aspect is the speed: predictions of AI agents outnumbering humans by 2025 seemed fantastical just months ago, yet major institutions are now actively planning for this reality.
The “teammate” framing is particularly significant. By positioning AI as colleagues rather than tools, companies are preparing their workforce for a psychological shift that will be necessary for effective collaboration. However, the trust deficit identified by JPMorgan’s research team reveals the central tension: organizations want AI’s efficiency but aren’t ready to fully relinquish human oversight.
The Cambridge study results are telling—AI excels at optimization and data-driven decisions but struggles with dynamic, real-time judgment. This suggests the future isn’t human replacement but rather a hybrid model where AI handles structured tasks while humans focus on adaptive decision-making, relationship management, and strategic thinking.
Why This Matters
This development represents a watershed moment for the financial services industry and the broader future of work. The deployment of AI agents as digital coworkers signals a fundamental shift from AI as a tool to AI as a collaborative workforce participant. For Wall Street specifically, where labor costs are substantial and junior talent pipelines are critical, AI agents could dramatically reshape career trajectories, compensation structures, and skill requirements.
The implications extend far beyond finance. If major banks successfully integrate AI agents into daily workflows, other industries will rapidly follow suit. The trust and verification challenges highlighted by JPMorgan’s research team underscore a universal tension: organizations must balance the efficiency gains of autonomous AI against the risks of errors and the need for human oversight.
The prediction of AI agents outnumbering humans by 2025 suggests an unprecedented acceleration in AI deployment. This raises critical questions about workforce displacement, the changing nature of entry-level positions, and how organizations will need to restructure teams, management practices, and training programs to accommodate hybrid human-AI workforces. The financial sector’s experience will likely serve as a blueprint for AI agent integration across the global economy.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources:
Recommended Reading
Related Stories
- Sam Altman’s Bold AI Predictions: AGI, Jobs, and the Future by 2025
- The Impact of AI on Software Engineering Jobs and Market Outlook
- Perplexity CEO Predicts AI Will Automate Two White-Collar Roles by 2025
- How Companies Can Use AI to Meet Their Operational and Financial Goals
- Mistral AI Launches Le Chat Assistant for Consumers and Enterprise
Source: https://www.businessinsider.com/how-banks-using-agents-generative-ai-fintechs-2024-11